Scholarlink

Objectives <p>The performance of fully automated deep learning-based models for the detection and segmentation of lymph nodes (LNs) on full- and simulated reduced-dose CT was validated.</p> Methods <p>A total of 15,341 LNs were annotated in 151 patient CTs (age 52 ± 14 years, 87 males) from the public TCIA NIH CT Lymph Nodes dataset. Two 3D nnU-Net models were trained on 90 CT scans: (1) only full dose CTs (NoAugmentation), and (2) both full- and reduced-dose CTs (Augmentation). Dose reduction from 75% to 5% of the full-dose was simulated using a noise-addition tool. Performance was validated on the remaining 61 CTs and an external TCIA Mediastinal LNQ dataset (120 CTs, 64 females).</p> Results <p>On 61 full-dose CTs, the Augmentation model detected all LNs with 67.3% precision and 84.6% sensitivity. For all LNs and large nodes (short axis diameter ≥ 8&#xa0;mm), Dice Similarity Coefficient (DSC) was 0.83 ± 0.07 and 0.80 ± 0.14, while Hausdorff Distance (HD) error was 1.47 ± 0.91&#xa0;mm and 3.2 ± 2.28&#xa0;mm, respectively. Performance decreased with dose reduction (<i>p</i> &lt; 0.01), reaching 73.8% detection sensitivity and 0.75 DSC at 5% dose. Statistically significant differences between Augmentation vs. NoAugmentation models were seen for all nodes (<i>p</i> &lt; 0.001) and small nodes (<i>p</i> &lt; 0.05) at 10% and 5% doses. On the external LNQ dataset, the Augmentation model attained a DSC of 0.76 ± 0.12 and HD of 4.7 ± 3.23 (<i>p</i> &lt; 0.01) for all LNs.</p> Conclusion <p>Degraded image quality impacted nodal delineation on reduced-dose CT. Performance improved when a model trained on both full- and reduced-dose CTs was used.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Analysis of deep learning-based segmentation of lymph nodes on full-dose and reduced-dose body CT

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月18日

查看全文
  • Lindsey H. Bloom,
  • Tejas Sudharshan Mathai,
  • Bohan Liu,
  • Brandon Khoury,
  • Naiya Patel,
  • Olivia Wei,
  • Praveen T. S. Balamuralikrishna,
  • Benjamin Hou,
  • Justin Solomon,
  • Darko Pucar,
  • Ehsan Samei,
  • Elizabeth C. Jones,
  • Ronald M. Summers

摘要

Objectives

The performance of fully automated deep learning-based models for the detection and segmentation of lymph nodes (LNs) on full- and simulated reduced-dose CT was validated.

Methods

A total of 15,341 LNs were annotated in 151 patient CTs (age 52 ± 14 years, 87 males) from the public TCIA NIH CT Lymph Nodes dataset. Two 3D nnU-Net models were trained on 90 CT scans: (1) only full dose CTs (NoAugmentation), and (2) both full- and reduced-dose CTs (Augmentation). Dose reduction from 75% to 5% of the full-dose was simulated using a noise-addition tool. Performance was validated on the remaining 61 CTs and an external TCIA Mediastinal LNQ dataset (120 CTs, 64 females).

Results

On 61 full-dose CTs, the Augmentation model detected all LNs with 67.3% precision and 84.6% sensitivity. For all LNs and large nodes (short axis diameter ≥ 8 mm), Dice Similarity Coefficient (DSC) was 0.83 ± 0.07 and 0.80 ± 0.14, while Hausdorff Distance (HD) error was 1.47 ± 0.91 mm and 3.2 ± 2.28 mm, respectively. Performance decreased with dose reduction (p < 0.01), reaching 73.8% detection sensitivity and 0.75 DSC at 5% dose. Statistically significant differences between Augmentation vs. NoAugmentation models were seen for all nodes (p < 0.001) and small nodes (p < 0.05) at 10% and 5% doses. On the external LNQ dataset, the Augmentation model attained a DSC of 0.76 ± 0.12 and HD of 4.7 ± 3.23 (p < 0.01) for all LNs.

Conclusion

Degraded image quality impacted nodal delineation on reduced-dose CT. Performance improved when a model trained on both full- and reduced-dose CTs was used.